CLI for DeepVista — chat, notes, recipes, and memory from your terminal.
Project description
deepvista-cli
CLI for DeepVista — chat, notes, recipes, and vistabase from your terminal. Designed for both humans and AI agents.
Table of Contents
- For AI Agents
- Install
- Uninstall
- Terminal UI (TUI)
- Authentication
- Profiles
- Commands
- Global Flags
- Output
- Exit Codes
- Environment Variables
- Files
- See Also
For AI Agents
Claude Code
Cursor
OpenCode
and any agent that supports skills
The key idea: install once, then talk to your agent. The agent handles authentication and all commands on your behalf.
Install
Run this in your terminal — no Node, no extra tools required beyond a Python package manager and either git or curl:
curl -sSL https://raw.githubusercontent.com/DeepVista-AI/deepvista-cli/main/install.sh | bash
The script:
- Installs the
deepvistaCLI (auto-detectsuv,pipx, orpip) - Copies skills into your agent's skills directory (auto-detects Claude Code, OpenCode, Cursor, and others)
| Skill | What it teaches your agent |
|---|---|
deepvista-shared |
Auth, profiles, global flags, security rules |
deepvista-vistabase |
Knowledge cards — search, read, create, update |
deepvista-notes |
Note capture and management |
deepvista-vistabook |
Run structured AI recipes/workflows |
deepvista-vistabase |
View and search implicit vistabase context |
deepvista-chat |
Conversational AI agent |
deepvista-persona-knowledge-worker |
Daily knowledge workflow patterns |
deepvista-recipe-research-to-vistabook |
Search → synthesize → run workflow |
deepvista-recipe-export-knowledge-as-skills |
Turn your knowledge into installable skills |
deepvista-recipe-analyze-notes |
Analyze, summarize, and find patterns across notes |
Get Started
Open your agent and paste:
Load skills: deepvista-shared deepvista-notes deepvista-vistabase
Help me get started with DeepVista. Walk me through logging in.
Your agent will:
- Open the browser login page
- Guide you through pasting the auth code
- Confirm you're logged in with
deepvista auth status
Claude Code tip: If you're prompted to confirm skill file reads, run once:
claude config set allowedPaths "~/.claude/skills" --global
Auto-Capture Notes
The install script automatically enables auto-capture in every detected agent — no manual setup required.
| Agent | Config file written |
|---|---|
| Claude Code | ~/.claude/CLAUDE.md |
| Cursor | ~/.cursor/rules |
| OpenCode | ~/.opencode/AGENTS.md |
Once installed, your agent will silently save facts, decisions, insights, and action items to your DeepVista knowledge base as you work — no manual invocation needed. The install is idempotent: re-running it won't duplicate the block.
To remove auto-capture, delete the <!-- deepvista-auto-capture --> block from the relevant config file, or run the uninstall script (see Uninstall).
Use Cases
Capture Insights from Podcasts
Build a searchable knowledge base from founder interviews.
After listening to an episode, ask your agent:
Load skills: deepvista-shared deepvista-notes
I just listened to the Lenny's Podcast episode with Brian Chesky about founder mode.
Here are my notes: ...
Save this to my knowledge base as a note titled "Brian Chesky — Founder Mode".
Search and retrieve across everything you've captured:
Load skills: deepvista-shared deepvista-vistabase
Search my knowledge base for everything I've captured about founder mindset and obsession.
Show me the top 5 results.
Research and Synthesize with a Recipe
Once you've captured 10–20 interviews, ask your agent to synthesize patterns:
Load skills: deepvista-shared deepvista-vistabase deepvista-vistabook deepvista-recipe-research-to-vistabook
Search for cards about growth and early-stage execution, then find my Research Synthesis
recipe and run it focused on: what separates high-growth founders, common 0→1 mistakes.
Show me what you find before running.
The agent searches, reads your notes, finds the Recipe, confirms with you, then streams the run live.
Build a Founder Playbook
Load skills: deepvista-shared deepvista-vistabase deepvista-vistabook
I have a new startup idea to evaluate. Idea: [your idea].
1. Search my knowledge base for any idea validation frameworks
2. Find my idea evaluation Recipe
3. Run it with the above context — show me the Recipe first
Export your playbook as a reusable skill:
Load skills: deepvista-shared deepvista-vistabook deepvista-recipe-export-knowledge-as-skills
Export my founder playbook Recipe as a SKILL.md file so I can share it with my team.
Recipe Patterns
Build these workflows in the DeepVista web app, then invoke them through your agent:
| Recipe | Prompt to invoke |
|---|---|
| Research synthesis | "Search my KB for [topic] and run my Research Synthesis Recipe" |
| Idea evaluation | "Evaluate this idea against my founder frameworks: [idea]" |
| Weekly review | "Run my weekly review — surface pinned cards and capture this week's key learnings" |
| Interview debrief | "I just finished a user interview. Run my Interview Debrief Recipe with these notes: [notes]" |
| Decision memo | "Help me think through this decision using my knowledge base: [decision]" |
| Competitive analysis | "Run a competitive analysis on [company/space] using everything I've captured" |
Install
From PyPI
pip install deepvista-cli
pipx install deepvista-cli
uv tool install deepvista-cli
With Terminal UI support
pip install 'deepvista-cli[ui]'
uv tool install 'deepvista-cli[ui]'
Directly from GitHub
pip install git+https://github.com/DeepVista-AI/deepvista-cli.git
uv tool install git+https://github.com/DeepVista-AI/deepvista-cli.git
For development (from this repo)
uv sync
uv pip install -e '.[ui]' # include TUI
uv run deepvista --help
Uninstall
curl -sSL https://raw.githubusercontent.com/DeepVista-AI/deepvista-cli/main/uninstall.sh | bash
The script:
- Uninstalls the
deepvistaCLI - Removes all DeepVista skills from detected agent skill directories
- Removes the auto-capture block from agent config files
Terminal UI (TUI)
DeepVista includes a full terminal UI with four panels matching the product's core modules.
Launch
deepvista ui
Requires the [ui] optional dependency:
pip install 'deepvista-cli[ui]'
Panels
| Panel | Key | Description |
|---|---|---|
| Chat | 1 |
Converse with the DeepVista AI agent. Select past sessions from the sidebar or start a new one. Responses stream in real time. |
| Notes | 2 |
Browse and search your explicit knowledge (notes). Click any note to read its full content. |
| Recipes | 3 |
List all Recipes (structured workflows). Select one to view its details and run it with the ▶ Run Recipe button — output streams live. |
| Memory | 4 |
Read-only view of implicit context accumulated from your Chat sessions. Searchable. |
Keyboard Shortcuts
| Key | Action |
|---|---|
1 – 4 |
Switch between Chat / Notes / Recipes / Memory |
r |
Refresh the active panel |
q |
Quit |
Enter |
Send message (in Chat input) |
Authentication
Browser login (default)
deepvista auth login
Opens your browser to DeepVista's login page. If you're already logged in, the CLI authenticates automatically via a localhost callback.
Non-interactive / headless
Visit the web app at /cli, sign in, and paste the code shown:
deepvista auth login --code XXXX-XXXX
Check / clear auth
deepvista auth status
deepvista auth logout
Profiles
Profiles store api_url so you don't need env vars for each environment.
# Create profiles
deepvista config set local --api-url http://localhost:8080
deepvista config set staging --api-url https://api-staging.deepvista.ai
# Use a profile
deepvista --profile local card list
deepvista --profile staging chat +send "hello" --new
# Manage
deepvista config list
deepvista config show local
deepvista config delete old
Resolution order (first wins): CLI flags → named profile → built-in default
Commands
The CLI uses four resources:
card Knowledge cards (all types)
recipe Executable workflows
vistabase Implicit context from Chat (read-only)
chat Conversational AI agent
Backward compatibility:
deepvista memoryis a deprecated alias fordeepvista vistabase.
card — Knowledge cards
# List cards
deepvista card list [--type TYPE] [--status pinned|archived|normal] [--limit N] [--page N]
# Get / create / update / delete
deepvista card get <card_id>
deepvista card create --type TYPE --title "Title" [--content "..."] [--tags '["t1"]']
deepvista card update <card_id> [--title "..."] [--content "..."] [--status pinned|archived]
deepvista card delete <card_id>
# Search (hybrid vector + keyword)
deepvista card +search "query text" [--type TYPE] [--limit 10]
# Find similar cards
deepvista card +similar <card_id>
# Pin / archive
deepvista card +pin <card_id>
deepvista card +archive <card_id>
Card types: person, organization, message, todo, topic, keypoint, file, note, recipe, recipe_run
recipe — Executable workflows
deepvista recipe list [--limit N] [--page N]
deepvista recipe get <recipe_id>
deepvista recipe run <recipe_id> [--input "context"]
deepvista recipe status <run_chat_id>
deepvista recipe export <recipe_id> --format skill
recipe run streams NDJSON as the agent works through the checklist.
vistabase — Implicit context
Vistabase is automatically accumulated from Chat. It is read-only — updates happen through conversation.
deepvista vistabase show [--limit N]
deepvista vistabase search "query text" [--limit N]
deepvista memoryis a deprecated alias that works identically.
chat — AI agent
deepvista chat sessions [--limit N] [--offset N] [--search "query"]
deepvista chat get <chat_id>
deepvista chat delete <chat_id>
deepvista chat +send "your message" [--chat-id ID] [--new]
chat +send streams NDJSON as the agent responds.
notes — Quick note management (shorthand)
notes is a convenience alias for card commands filtered to type=note.
deepvista notes list [--limit N]
deepvista notes get <note_id>
deepvista notes create --title "Title" [--content "..."]
deepvista notes update <note_id> [--title "..."] [--content "..."]
deepvista notes delete <note_id>
deepvista notes +quick "Quick note from a single line of text"
Global Flags
Global flags must come before the resource name:
# Correct
deepvista --profile local card list
# Wrong
deepvista card list --profile local
| Flag | Default | Description |
|---|---|---|
--format json|table |
json |
Output format |
--verbose |
off | Show HTTP request/response details |
--dry-run |
off | Show what would be sent, don't execute |
--api-url URL |
https://api.deepvista.ai |
Override backend URL |
--profile NAME |
default |
Use a named config profile |
--version |
Show version and exit |
Output
- JSON (default): structured JSON to stdout — agents parse this.
- Table:
--format tablefor human-readable output. - Errors:
{"error": {"code": N, "message": "...", "detail": "..."}}on stderr. - Streaming: NDJSON for
chat +sendandrecipe run(one JSON object per line).
Exit Codes
| Code | Meaning |
|---|---|
| 0 | Success |
| 1 | API error |
| 2 | Auth error (not logged in / token expired) |
| 3 | Validation error (bad arguments) |
| 4 | Network error |
| 5 | Internal error |
Environment Variables
| Variable | Description |
|---|---|
DEEPVISTA_CONFIG_DIR |
Config directory (default: ~/.config/deepvista) |
Files
deepvista-cli/
├── deepvista_cli/
│ ├── main.py # CLI entry point (five resources)
│ ├── config.py # Config + profiles
│ ├── auth/ # Login, token storage, callback server
│ ├── client/ # HTTP client, SSE streaming
│ ├── commands/
│ │ ├── card.py # Knowledge cards
│ │ ├── recipe.py # Executable workflows
│ │ ├── memory.py # Implicit memory context
│ │ ├── chat.py # Conversational agent
│ │ ├── notes.py # Notes (alias for card --type note)
│ │ ├── auth.py
│ │ └── config.py
│ ├── tui/
│ │ └── app.py # Terminal UI (requires [ui] extra)
│ └── output/ # JSON + table formatters
└── skills/ # SKILL.md files for agent integration
├── deepvista-shared/
├── deepvista-vistabase/ # card commands
├── deepvista-vistabook/ # recipe commands
├── deepvista-vistabase/ # vistabase commands
├── deepvista-notes/
├── deepvista-chat/
├── deepvista-persona-knowledge-worker/
├── deepvista-recipe-analyze-notes/
├── deepvista-recipe-export-knowledge-as-skills/
└── deepvista-recipe-research-to-vistabook/
See Also
- skills/ — SKILL.md files for agent integration
- deepvista.ai — Web app to build and manage Recipes
- PyPI — Package releases
- GitHub Issues — Bug reports and feature requests
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file deepvista_cli-0.1.0a11.tar.gz.
File metadata
- Download URL: deepvista_cli-0.1.0a11.tar.gz
- Upload date:
- Size: 69.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f4fcf0e1863e68dfcd3826b0b7dfd0a39e69a0873fad5760f0d1534ed614863a
|
|
| MD5 |
45e73e540039f867f80dc200a665d251
|
|
| BLAKE2b-256 |
e793bb6421ce2f12322cbcb4d08878c0cab93f2e1574e2296941659cf3bf774a
|
Provenance
The following attestation bundles were made for deepvista_cli-0.1.0a11.tar.gz:
Publisher:
publish.yml on DeepVista-AI/deepvista-cli
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
deepvista_cli-0.1.0a11.tar.gz -
Subject digest:
f4fcf0e1863e68dfcd3826b0b7dfd0a39e69a0873fad5760f0d1534ed614863a - Sigstore transparency entry: 1256655638
- Sigstore integration time:
-
Permalink:
DeepVista-AI/deepvista-cli@6b921106a34ff8ae2745d0987357308a2a08f185 -
Branch / Tag:
refs/tags/v0.1.0a11 - Owner: https://github.com/DeepVista-AI
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@6b921106a34ff8ae2745d0987357308a2a08f185 -
Trigger Event:
push
-
Statement type:
File details
Details for the file deepvista_cli-0.1.0a11-py3-none-any.whl.
File metadata
- Download URL: deepvista_cli-0.1.0a11-py3-none-any.whl
- Upload date:
- Size: 50.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f43a62d9328c77907eb7ebdcdc6b307665209ee559907bf7d128cf7b3888d038
|
|
| MD5 |
e385c473b11457917480c5df2b96f8de
|
|
| BLAKE2b-256 |
ebdb016d561458b0c5a90a6ade56d093e2dbe30230ee1f7a54c6223994030584
|
Provenance
The following attestation bundles were made for deepvista_cli-0.1.0a11-py3-none-any.whl:
Publisher:
publish.yml on DeepVista-AI/deepvista-cli
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
deepvista_cli-0.1.0a11-py3-none-any.whl -
Subject digest:
f43a62d9328c77907eb7ebdcdc6b307665209ee559907bf7d128cf7b3888d038 - Sigstore transparency entry: 1256655762
- Sigstore integration time:
-
Permalink:
DeepVista-AI/deepvista-cli@6b921106a34ff8ae2745d0987357308a2a08f185 -
Branch / Tag:
refs/tags/v0.1.0a11 - Owner: https://github.com/DeepVista-AI
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@6b921106a34ff8ae2745d0987357308a2a08f185 -
Trigger Event:
push
-
Statement type: